Editor Workflow

Workflow Overview

1. Design - Plan your AI behavior
2. Create - Build the tree in the editor
3. Configure - Set up blackboard and properties
4. Test - Debug in the editor
5. Export - Generate runtime assets
6. Integrate - Use in your game

Design Phase

Before opening the editor, plan your AI:

Enemy AI Design:
├── If health > 50%
│   ├── Find target
│   ├── Move to target
│   └── Attack
└── Else
    └── Flee to safety

Create Phase

Translate your design to nodes:

1. Start with a Selector (main decision)
2. Add Sequences for each branch
3. Add Conditions and Actions
4. Configure node properties

Configure Phase

Blackboard Setup

Define variables your tree needs:

Variable	Type	Default	Description
health	number	100	Current health
target	Entity	null	Attack target
homePosition	Vector2	(0,0)	Safe position

Property Bindings

Connect node properties to blackboard:

Attack Node:
  damage: @blackboard.attackPower
  target: @blackboard.target

Test Phase

1. Click Play in the editor
2. Watch node execution
3. Monitor blackboard values
4. Step through execution
5. Fix issues and repeat

Export Phase

Export your tree for runtime use:

// The exported JSON can be loaded at runtime
const treeData = await loadBehaviorTree('assets/enemy-ai.json');

Integration Phase

import { BehaviorTreeLoader, BehaviorTreeStarter } from '@esengine/behavior-tree';

// Load exported tree
const treeData = await BehaviorTreeLoader.load('enemy-ai.json');

// Attach to entity
const enemy = scene.createEntity('Enemy');
BehaviorTreeStarter.start(enemy, treeData);